Intelsat is devising a transformational business plan for after it emerges from bankruptcy restructuring later this year, including a big bet on software-defined satellites and potentially its own low-Earth-orbit broadband constellation.
The operator, which has been in Chapter 11 bankruptcy protection for nearly a year and a half, issued a request for proposals (RFP) at the end of July for 10 satellites that could be reconfigured in-orbit for changing mission needs.
What role will software-defined satellites play for Intelsat after it emerges from restructuring?
Software-defined satellites are an integral part of our strategy. We have ordered two of them so far, in the factory today, and we have just launched an RFP for 10 of them. We’re not going light on this. This is a major investment in technology.
Is there a deadline for this RFP?
We just launched it, and the time period is going to be in the next three to five years — it’s not too extended.
We don’t need the 10 to cover the globe. Our first two, plus what we have today, give us pretty much global coverage. The rest is all adding capacity where it’s needed. So it’s not from a coverage point of view, but more of a densification element. The beauty about those things is, as time elapses, you need less and less time from contract to orbit. So we can get them up in a very quick manner.
What else can you say about what these 10 satellites will do — I suppose them being software-defined means you can change that on the fly?
Yes, Intelsat has the richest portfolio of slots in the industry, so it depends on the mission. We obviously have a plan right now for our first two and where we’re going to put those, but for the rest we’re just going to plug the holes or add capability as needed.
Does it just make sense then to bulk-buy as many software-defined satellites as you can, because of the supply chain issues currently facing the industry?
Partly, but also we have a business strategy coming out of this restructuring that is growth-orientated. In order to deliver that growth, you need the assets behind it. So we’re designing a network that can more than satisfy our needs. When you look at commercial aviation connectivity, for example, you realize that today’s take rates on airplanes is very low. You’re talking about 8-10% take rates on airplanes, because the Wi-Fi services have traditionally been low quality or expensive, but that model is changing to where a lot of airlines now are offering this as a freemium, ad-supported or content-supported model.
When you move to this environment, the take rate is going to jump dramatically. So the capacity that will be required on airplanes will increase by orders of magnitude. Then to keep up, because you want to provide the right quality, you’ll need a lot of capacity. That drives the requirement for up to 10 satellites.
So it’s about the traditional markets, but also markets that when they grow they require so much more capacity. Similarly in the cruise market. You used to go on cruises to disconnect, but now their advertising is based on the fact that you can have internet onboard.
This is also great news for the GEO manufacturing market.
Yes, today we have 10 satellites in the factory. Seven C-band satellites, the two software-defined satellites, and an advanced version of our [high-throughput platform] Epic.
I don’t even know if a factory a year or two ago even had 10 satellites combined. I think in many ways we have revived some manufacturers that may have gone bankrupt otherwise. You’re absolutely right, I think there’s a rejuvenation within the GEO market. It’s centered around the traditional C-band satellites that will continue to be there for a long time, but also I think you’ll see a lot more software-defined satellites.
I imagine you’re going to want to spread these 10 satellites around different manufacturers?
When do you expect to decide on all of this?
I think within the first quarter of next year, as we emerge from the restructuring and finalize our business plan.
How do these software-defined satellites fit within your existing GEO fleet?
Our view is that we need to have a system that really has multiple layers to it in GEO for serving different applications. You have your traditional C-band satellites that serve our media customers — and our media businesses are still 42% of our business today, and then you have wide-beam satellites that serve applications like intelligence, surveillance and reconnaissance, or ISR. Then we have Epic satellites with high throughput that serve our network business and mobility customers. On top of that, we want our software-defined satellites.
The idea is, within the GEO layer, we would be able to serve the customer with the most optimum capacity for the application that they are utilizing, at the location where they are.
Especially when you talk about commercial aviation, you want a lot of capacity density in some areas, and we are able to do that with this multi-layer approach.
The second part of the strategy is a multi-band capability. Traditionally, most of our business has been in Kuband, but we have also started offering Ka-band services. In the future, we see ourselves moving into Q, V and other bands as well. So it’s not only about Ku-band, although today we hold a lot of slots and frequencies in this band.
What about opportunities Intelsat sees in a nongeostationary orbit (NGSO)?
The third element of our strategy is multi-orbit, because as much as we believe GEO is fundamental to our offering, we think that there is an advantage to an NGSO constellation that could provide capabilities that are mainly centered around latency and coverage. Especially when you start looking at the poles, which are becoming increasingly important both for government and commercial aviation customers. So we’re defining and designing our NGSO strategy today. We haven’t made any announcements yet. It’s no secret that we’re working in partnerships with some of the existing players, but we’re also looking at potentially building our own.
We do not believe that a stand-alone NGSO system is ever going to make money. We don’t believe that you can get a return on investment. We believe that people who do it, do it for other reasons that’s of value to them for other parts of their businesses.
We’ve looked into this quite a bit and we’ve explored that enough to know that you just cannot monetize a system like this — but we believe that if you put that as a complement to a GEO constellation, it starts making sense.
Firstly, because you don’t have to build something that is worldwide. You can build it to cover where your gaps are. Secondly, you’re building it on top of an existing platform of revenues and customer base.
The biggest problem we believe in NGSO is the fact that you have to build the system, and then you have a very short window of time in which you have to monetize the whole thing so you can replenish the system, because every five years you need new satellites, right? When you have an existing customer base and you’re just upselling them services or bundling, it’s different from when you’re trying to build a new customer base. From that point of view, we believe that our NGSO strategy is probably the right one.
So this is what we’re building. It’s a multi-orbit, multi-layer and multi-band strategy from a space segment.
Intelsat entering the NGSO market directly would be big news.
We haven’t made any announcements on this. We’re exploring today the possibility of us building, as opposed to just partnering. There are more things to say about that in the future.
When do you expect to make a decision on that?
I think early next year we’ll have probably formulated our plans and made them public. You must have seen Inmarsat recently talking about adding a LEO element to their GEO constellation? I think a lot of people understand that the combination of the two is useful. Whether people pull it off or not, that’s a different story.
And this is all part of a new chapter in Intelsat’s story as it seeks to emerge with a fresh balance sheet after restructuring?
We are quite diversified as a company, if you think about it, we serve all of the sectors. This diversification has, by the way, been extremely helpful for us in a COVID-19 environment, where you had some sectors that got hit, but others that did really well.
But how do you serve multiple sectors?
If you look at some of the NGSO designs, for example, some of our competitors have very inflexible designs or they have like one flavor of the space segment — you can never serve multiple sectors with one flavor of space segment. You can do it, but you’re just doing it in a very non-optimal way. For us, what we want to do is use the different types of space segment capability that we’re bringing to optimally serve different verticals with each. For that, of course, you need scale.
You spoke about how having an existing customer base is an essential foundation. Is there also an advantage to not being a first-mover? Ground antennas have been a big issue for the LEO broadband pioneers. Are you expecting costs to have come down by the time Intelsat arrives?
You’re 100% right. Three years ago, no one was talking about multi-orbit terminals. Everyone was talking about just trying to get one terminal in place and that wasn’t easy. So we’re starting from the gate talking about multi-orbit terminals. The ground is where the other part of the innovation is and where our focus is — having infrastructure that is completely 5G-capable.
We’re starting with a core network that is 5G, so that the interoperability with our customers becomes seamless. That’s very, very important. We’re also looking into potentially using a 5G waveform in the future. We’re taking it to the point where terrestrial technology is really at the heart of what we use to communicate. Not just in the back office, but the service will be 5G-capable. That’s what we’re shooting for.
Enabling handsets that consumers are already using to connect to satellites as well as terrestrial networks?
There are two parts to this. One, you have to be able to close the link with the satellite asset. With a GEO, you would never be able to close the link directly to the handset. But we’re not just talking about GEO now. We’re talking multiple layers of capability, even probably to the point of having high altitude platforms [such as autonomous airships], and they are things that can communicate directly with the device.
So you have that link, but then you also have the technology — the waveform, what is that based on? We as a company have been driving Release 17 of the 3GPP standard. That defines how satellite integrates seamlessly within a terrestrial environment. That is something we’re working heavily on, plus having our own cloud services as well.
In addition to this, we’re heavy on the concept of virtualization. Look at COVID, a lot of people struggled not because the bandwidth requirements weren’t there, but because they couldn’t get engineers on sites to upgrade equipment. When you’re in a world of virtualization, you don’t have that problem, and you can rely more on software as opposed to a business of inefficient hardware that you have to keep replicating and produce.
Open architecture has a role to play in this?
We’ve always had an open architecture as opposed to vertically integrated, closed technology. What we’re doing today is we’re opening the door for like-minded operators in the industry to join us in some form of alliance program where they can bring their capabilities, put them in the network with us and vice versa, and we can orchestrate all of that to the benefit of customers. So customers that work with us will not only have the benefit of having access to all of our space assets today, and the ones that are coming up, but also the assets of others that could be competitors on the service side, but we believe that we can have offerings that really make sense.
Doing this really builds scale, and this is a business of scale. The scale that we’re looking at for the future is really orders of magnitude higher than what it used to be.
So Intelsat’s view is one satellite operator will not be able to reach the scale to serve everyone everywhere, and partnerships are needed to patch a network together?
Absolutely, and that’s the right strategy. There are companies that obviously want to do things on their own, and they’re vertically integrated. I think they will struggle, and it will be difficult for them to be self-sustainable.
This interview has been edited for length and clarity.
This article originally appeared in the October 2021 issue of SpaceNews magazine.